[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122715-en":3,"doc-seo-122715-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":26,"seo_description":14,"update_tm":27,"read_time":28},122715,34359740700684,"Finn","https://ap-avatar.wpscdn.com/avatar/1f400023980c374ae676?_k=1777273430885731487",8,"Research & Report","Overcome Ethnic Discrimination with Unbiased Machine Learning for Facial Data Sets - research network and fair CNN training","AI-based prediction and recommender systems are widely deployed, yet acceptance of AI-enabled systems remains insufficiently studied. A survey of 559 respondents indicates that fairness, transparency, suitable consideration of personality traits, and efficient task performance shape acceptance. The work then targets biased facial annotations by introducing AntiDiscriminationNet (ADN): an unbiased attractiveness prediction network trained with synthetic images and weighted data for anti-discrimination assessments across ethnicities, including entropy-penalty techniques to reduce implicit bias.","# Overcome Ethnic Discrimination with Unbiased Machine Learning forFacial Data Sets*\n\nMichael Danner¹D²,Bakir Hadžic²,t,Robert Radloff²,Xueping Su³D,Leping Peng⁴,Thomas Weber²and Matthias Rätsch²DC  \n¹CVSSP,University of Surrey,Guildford,U.K.2ViSiR,Reutlingen University,Germany  \n3School of Electronics and Information,Xi'an Polytechnic University,China4Hunan University of Science and Technology,China  \nKeywords:Unbiased Machine Learning,Fairness,Trustworthy AI,Acceptance Research,Debiasing Training Data,Facial Data Sets,AI-Acceptance Analysis.  \nAbstract:AI-based prediction and recommender systems are widely used in various industry sectors.However,generalacceptance of AI-enabled systems is still widely uninvestigated.Therefore,firstly we conducted a survey with559 respondents.Findings suggested that AI-enabled systems should be fair,transparent,consider person-ality traits and perform tasks efficiently.Secondly,we developed a system for the Facial Beauty Prediction(FBP)benchmark that automatically evaluates facial attractiveness.As our previous experiments have proven,these results are usually highly correlated with human ratings.Consequently they also reflect human biasin annotations.An upcoming challenge for scientists is to provide training data and AI algorithms that canwithstand distorted information.In this work,we introduce AntiDiscriminationNet(ADN),a superior attrac-tiveness prediction network.We propose a new method to generate an unbiased convolutional neural network(CNN)to improve the fairness of machine learning in facial dataset.To train unbiased networks we generatesynthetic images and weight training data for anti-discrimination assessments towards different ethnicities.Additionally,we introduce an approach with entropy penalty terms to reduce the bias of our CNN.Our re-search provides insights in how to train and build fair machine learning models for facial image analysis byminimising implicit biases.Our AntiDiscriminationNet finally outperforms allcompetitors in the FBP bench-mark by achieving a Pearson correlation coefficient of PCC=0.9601.  \n## 1 INTRODUCTION\n\nbeen conducted to understand the evolutionary basisof beauty and determine the bias of attractiveness inthe job hiring process (Little et al.,2011;Chiang andSaw,2018).Companies desire an efficient and objec-tive recruitment process with the preferred outcomeof finding the best job candidates and stay compliantwith regulations and ethical aspects.Artificial intelli-gence has the potential to support these goals by min-imising the risk of bias in decision making in orderto be a relevant and trustworthy partner for humans inthe future.  \nIn recent years,the use of artificial intelligence hasproven to solve a wide spectrum of technical prob-lems.Especially in the high-tech sector and in knowl-edge intensive industries,machines and intelligent al-gorithms turned from clunky tools to sophisticatedsystems performing various complex tasks today(Ar-slan et al.,2021).In today's global war of talents,companies are hunting for the best employees withspecific requirements of skills and personal traits toachieve competitive advantage in their field(Grant,1991).In this context,a wide range of research has  \n### 1.1 Motivation\n\nIn 2016 Beauty.AI,a Hong-Kong based technologycompany,hosted the first international beauty contestjudged by artificial intelligence(beauty.ai,2016)butthe results were heavily biased,for example,againstdark-skinned subjects (Levin,2016).\"Machine learn-ing models are prone to biased decisions,due to bi-  \n464  \nases in data-sets”(Sharma et al.,2020).Biased train-ing data potentially leads to discriminatory models,asthe data sets are created by humans or derived fromhuman activities in the past,for example hiring algo-rithms(Bogen,2019).The purpose of Facial BeautyPrediction(FBP)research is to classify images mim-icking subjective human judgements.Investigationsrelated to machine perception in a ground-truth freesetting show that the data sour","cbCair5Zxl6WpuLT","https://ap.wps.com/l/cbCair5Zxl6WpuLT","pdf",3103559,1,"English","en",105,"# 1 INTRODUCTION\n## 1.1 Motivation\n## 1.2 Acceptance of AI","[{\"question\":\"What does the survey of 559 respondents reveal about AI system acceptance?\",\"answer\":\"Acceptance depends on fairness and transparency, consideration of personality traits, and efficient task performance.\"},{\"question\":\"What is AntiDiscriminationNet (ADN) used for?\",\"answer\":\"ADN is a facial attractiveness prediction network designed to improve fairness by reducing bias across different ethnicities.\"},{\"question\":\"How does the proposed method reduce bias in training facial machine learning models?\",\"answer\":\"It generates synthetic images, weights training data for anti-discrimination assessments, and applies entropy-penalty terms to lessen bias in the CNN.\"}]","Overcome Ethnic Discrimination with Unbiased Machine Learning for Facial Data Sets - research network and fair CNN training | PDF",1785812495,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":23,"language":22,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"overcome-ethnic-discrimination-with-unbiased-machine-learning-for-facial-data-sets-research-network-and-fair-cnn-training","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/overcome-ethnic-discrimination-with-unbiased-machine-learning-for-facial-data-sets-research-network-and-fair-cnn-training/122715/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":22,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What does the survey of 559 respondents reveal about AI system acceptance?","Question",{"text":74,"@type":75},"Acceptance depends on fairness and transparency, consideration of personality traits, and efficient task performance.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What is AntiDiscriminationNet (ADN) used for?",{"text":79,"@type":75},"ADN is a facial attractiveness prediction network designed to improve fairness by reducing bias across different ethnicities.",{"name":81,"@type":72,"acceptedAnswer":82},"How does the proposed method reduce bias in training facial machine learning models?",{"text":83,"@type":75},"It generates synthetic images, weights training data for anti-discrimination assessments, and applies entropy-penalty terms to lessen bias in the CNN.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]